# US 2021/0334697 A1 · Artificial Intelligence Recommendation System > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0334697 A1 - Document ID: US20210334697A1 - Status: Published application - Kind: A1 - Application number: 16/860,601 - Filed: 2020-04-28 - Published: 2021-10-28 - Earliest claimed: 2020-04-28 - Assignees: Optum Services (Ireland) Limited - Applicants: Optum Services (Ireland) Limited - Inventors: Darragh Hanley; Damian Kelly; Julie Zhu; Jonathan Lawrence Herke; Gregory J. Boss - Web page: https://patentgrep.com/patent/US20210334697A1 ## CPC - G06N 3/08: Learning methods - G06N 20/00: Machine learning - G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] - G06N 3/045: Combinations of networks - G06N 3/09: Supervised learning - G06N 5/02: Knowledge representation; Symbolic representation - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Various embodiments of the present disclosure facilitate recommendation prediction using machine learning. In one example, an embodiment provides for generating embeddings data related to one or more provider entities, predicting a set of provider entities for a patient entity based on a provider machine learning model, ranking provider entities in the set of provider entities to generate a ranked set of provider entities, and performing one or more actions to provide a recommendation for the patient entity based on the ranked set of provider entities. ## Claims ### Claim 1 A computer-implemented method for facilitating recommendation prediction using machine learning, the computer-implemented method comprising: generating embeddings data that encodes semantic relations among latent features for provider entities, wherein the embeddings data comprises at least an embedding vector that provides a vector representation for a provider entity; predicting a set of provider entities for a patient entity based on a provider machine learning model that is trained using the embeddings data, wherein a concatenation of user data for the patient entity and the embedding vector for the provider entity is provided to a machine learning network cell of the provider machine learning model, and wherein a machine learning network layer communicatively coupled to the machine learning network cell provides a prediction for the set of provider entities; ranking provider entities in the set of provider entities to generate a ranked set of provider entities; and performing one or more actions to provide a recommendation for the patient entity based on the ranked set of provider entities. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein the generating the embeddings data comprises generating the embeddings data based on one or more graph embedding techniques that transform a graph structure of the latent features into the embeddings data for the provider machine learning model. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein the generating the embeddings data comprises generating the embeddings data based on one or more factorization machines that transform a graph structure of the latent features into the embeddings data for the provider machine learning model. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein the ranking the provider entities comprises ranking the provider entities based on respective scores associated with the provider entities. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, wherein the ranking the provider entities comprises ranking the provider entities based on respective distance measurements associated with the provider entities. ### Claim 6 (depends on 1) The computer-implemented method of claim 1, further comprising: filtering the ranked set of provider entities to generate a filtered set of provider entities. ### Claim 7 (depends on 6) The computer-implemented method of claim 6, wherein the performing the one or more actions to provide the recommendation comprises performing the one or more actions to provide the recommendation for the patient entity based on the filtered set of provider entities. ### Claim 8 (depends on 1) The computer-implemented method of claim 1, further comprising: filtering the ranked set of provider entities based on one or more constraints associated with at least one of availability data for the set of provider entities, geographic location data for the set of provider entities, service type data for the set of provider entities, ratings data for the set of provider entities, and procedure data for the set of provider entities. ### Claim 9 An apparatus for facilitating recommendation prediction using machine learning, the apparatus comprising at least one processor and at least one memory including a computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to: generate embeddings data related to one or more provider entities that encodes semantic relations among latent features for provider entities, wherein the embeddings data comprises at least an embedding vector that provides a vector representation for a provider entity; predict a set of provider entities for a patient entity based on a provider machine learning model that is trained using the embeddings data, wherein a concatenation of user data for the patient entity and the embedding vector for the provider entity is provided to a machine learning network cell of the provider machine learning model, and wherein a machine learning network layer communicatively coupled to the machine learning network cell provides a prediction for the set of provider entities; rank provider entities in the set of provider entities to generate a ranked set of provider entities; and perform one or more actions to provide a recommendation for the patient entity based on the ranked set of provider entities. ### Claim 10 (depends on 9) The apparatus of claim 9, wherein the embeddings data is generated based on one or more graph embedding techniques that transform a graph structure of the latent features into the embeddings data for the provider machine learning model. ### Claim 11 (depends on 9) The apparatus of claim 9, wherein the embeddings data is generated based on one or more factorization machines that transform a graph structure of the latent features into the embeddings data for the provider machine learning model. ### Claim 12 (depends on 9) The apparatus of claim 9, wherein the provider entities are ranked based on respective scores associated with the provider entities. ### Claim 13 (depends on 9) The apparatus of claim 9, wherein the provider entities are ranked based on respective distance measurements associated with the provider entities. ### Claim 14 (depends on 9) The apparatus of claim 9, wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus to: filter the ranked set of provider entities to generate a filtered set of provider entities. ### Claim 15 (depends on 14) The apparatus of claim 14, wherein the recommendation for the patient entity is provided based on the filtered set of provider entities. ### Claim 16 A non-transitory computer storage medium comprising instructions for facilitating recommendation prediction using machine learning, the instructions being configured to cause one or more processors to at least perform operations configured to: generate embeddings data related to one or more provider entities that encodes semantic relations among latent features for provider entities, wherein the embeddings data comprises at least an embedding vector that provides a vector representation for a provider entity; predict a set of provider entities for a patient entity based on a provider machine learning model that is trained using the embeddings data, wherein a concatenation of user data for the patient entity and the embedding vector for the provider entity is provided to a machine learning network cell of the provider machine learning model, and wherein a machine learning network layer communicatively coupled to the machine learning network cell provides a prediction for the set of provider entities; rank provider entities in the set of provider entities to generate a ranked set of provider entities; and perform one or more actions to provide a recommendation for the patient entity based on the ranked set of provider entities. ### Claim 17 (depends on 16) The non-transitory computer storage medium of claim 16, wherein the embeddings data is generated based on one or more graph embedding techniques that transform a graph structure of the latent features into the embeddings data for the provider machine learning model. ### Claim 18 (depends on 16) The non-transitory computer storage medium of claim 16, wherein the embeddings data is generated based on one or more factorization machines that transform a graph structure of the latent features into the embeddings data for the provider machine learning model. ### Claim 19 (depends on 16) The non-transitory computer storage medium of claim 16, wherein the operations are further configured to: filter the ranked set of provider entities to generate a filtered set of provider entities. ### Claim 20 (depends on 19) The non-transitory computer storage medium of claim 19, wherein the recommendation for the patient entity is provided based on the filtered set of provider entities.